Alerts and notifications
Alerts is the record of anomaly and budget threshold alerts raised for spend you can see. Use it when a notification arrived earlier and you need to know what fired, when it fired, and what happened afterward.
The alert bell is the short list. See all alerts opens this full history.
Alerts, anomalies and budgets answer different questions
Section titled “Alerts, anomalies and budgets answer different questions”Alerts tells you what fired. It keeps the record even when the situation later changes.
The Anomaly Inbox is the work queue. It shows the cost spikes that need investigation and provides the actions for dealing with them.
Budgets & Forecasts shows where each budget stands now. A threshold alert can remain in this history after spend has moved back below the threshold, because the alert still happened.
Use Open anomaly or Open budgets on a row to reach the relevant screen.
New does not mean unresolved
Section titled “New does not mean unresolved”The bell badge is broader than this screen: it combines unseen alerts that still need attention with unread assignment notifications. Open the bell to review both. Closing its popover marks the current alerts as read and clears the alert part of the count. Reading an alert does not close an anomaly or change its status.
The full screen keeps read alerts. For dismissed, postponed and completed anomalies, it also keeps the alert and says what happened:
- A dismissed alert names who dismissed it and includes their reason.
- A postponed alert names who postponed it and the date it returns.
- An alert closes automatically when its spend returns to the expected baseline.
You do not close an anomaly by hand. CloudMonitor closes it once the spend returns to normal. Dismiss an expected spike with a reason, or postpone one you plan to handle later.
Budget alerts have no triage action on this screen. Open Budgets & Forecasts to find the budget’s current position and change its settings. Budgets are alert-only: crossing a threshold does not cap or stop spending.
New recommendations arrive in Teams, Slack and email
Section titled “New recommendations arrive in Teams, Slack and email”You do not need to open CloudMonitor to hear about a new recommendation. When one appears, whether it comes from Azure Advisor, a CloudMonitor rule or an unused Microsoft 365 Copilot seat, CloudMonitor sends one message to each destination set up for its cost group, or to the organization’s default destination. How to set Teams, Slack and email alerts shows where destinations are set.
A message lists the recommendations with the highest estimated saving first. Each one shows what to do, the resource it applies to and the yearly saving. Recommendations with no saving estimate are folded into one closing line, and the message says how many more there are when the list is long. It has one button, Open in CloudMonitor. The button opens Savings, or the Copilot seats screen when every recommendation in the message is an unused Copilot seat. When the link comes from the Fabric portal, the app opens on its home page.
A Copilot seat section names each person, says how long ago they last used Copilot and states the monthly saving once a seat price is entered in Settings. CloudMonitor cannot change licenses, so the message points to the Microsoft 365 admin center. The names appear only in the message.
Each recommendation is announced once. One you have dismissed, postponed or completed is not announced. The first message after installation lists the highest-saving recommendations that were already open and marks them as existing.
When nobody else would receive it
Section titled “When nobody else would receive it”A new installation often has no destination and no one in the Unallocated cost group. In that case the Organization Owner receives the message by email, for recommendations and for anomalies. A cost group that has a destination of its own, Unallocated included, keeps using it, and the Organization Owner receives nothing extra. The owner’s own notification choice applies: someone who turned their alerts off receives nothing.
Narrowing the history
Section titled “Narrowing the history”Use All, Anomalies or Budgets to show both alert types or one type. Each choice includes a count, and alerts stay newest first.
The top filter bar does not narrow this screen. An alert is a record of something that fired, not a report recalculated for the selected period. Changing Period, provider, cost group, virtual tags or cost basis therefore does not make alerts disappear.
Access still applies. You see alerts for cost groups you are allowed to read, even though the top-bar Cost group choice is not used here.
Empty and unavailable are different
Section titled “Empty and unavailable are different”No alerts have been raised. No action is needed. Anomalies and budget thresholds appear here when they fire. means the feed was read successfully and there is no history yet.
If CloudMonitor cannot read the alert history, the screen says that instead and reports the problem at the top of the app. It never presents a failed read as an empty history.